Mending Method for Partially Missing Data Using Cubic Spline Parameter Functions

This article first discusses the importance of complete data for data mining and analyzes the limits of some methods of solving missing data at present,and proposes the mending method for missing data using cubic parameter spline function,which is easy to use and do not need a lot of sample training. The method is compared with the principal component analytical method and the column average method,and it is ensured that cubic parameter spline function is close to the principal component analytical method. If the structure of cubic parameter spline function is further perfected,it will obtain the better effect.